design of a pneumatic robotic arm for solar cell tester

Design of a Pneumatic Robotic Arm for Solar Cell Tester

System by Using PLC Controller

1Yousif I. Al Mashhadany*, 2Nasrudin Abd Rahim

2 University of Malaya Power Energy Dedicated Advanced Centre (UMPEDAC),

Level 4, Wisma R&D, University of Malaya, Jalan Pantai Baharu,

59990 Kuala Lumpur, Malaysia, 1Electrical Engineering Department, College of Engineering, University of Al-Anbar, Al-Anbar, Iraq

Abstract

Solar cell testers sort photovoltaic cells according to their

electrical performance, tested under simulated sunlight. A

variety of testers exist, but they all face a common

challenge of handling cells that are very small and thin,

which makes it difficult to transport the cells from the

conveyer to the storage box. This paper presents a new

design for a handling robot with vacuum end-effectors,

which uses a PLC controller to govern the movement of

the cells and the testing process. The design applies to

solar cell testers for monocrystalline, polycrystalline,

cadmium telluride (CdTe), and copper indium diselenide

(CIS) cells. Each cell is tested for efficiency and

categorized accordingly into four groups (A to D). A

Virtual Reality (VR) model was built to simulate the

system, keeping in mind real world constraints. Two

photoelectric sensors were used to make detections for

both the testing process and the robot movement. The PLC

controller guides the trajectory of the robot according to

the results of the efficiency testing. It was seen that the

system worked very well, with the testing process and the

robot movement interacting smoothly. The robot trajectory

was seen to be highly accurate, and the pick and place

operations were done with great precision.

Keywords: Handling robot, Solar cell tester, Virtual reality,

photoelectric sensor.

1. Introduction

The testing of solar cells both in the laboratory and for

commercial uses, solar simulators are used. The simulators

employ a single light such as that from a xenon arc lamp to

produce a beam of collimated light that matches a

reference solar spectrum. Despite some advantageous

features for simulating a solar spectrum, the beam from a

xenon arc lamp cannot be directly used. This is firstly

because a considerable spectral range, especially between

0.8m and 1.2m, is covered in the emission lines of the

lamp output. Secondly, the richness of the Ultraviolet (UV)

region in the lamp output is an issue. To get around this

problem, there are proprietary notch filters that decrease

IR and UV components of the output, which is why many

types of optical filters are required to obtain a useful

spectrum. A water cavity is sometimes used to enhance

spectral matching. Still, a number of disadvantages of the

xenon lamp remain [1- 6]:

(i) The wavelength of the peak of the solar spectrum

and the xenon spectrum do not coincide.

(ii) Intense emission lines remain in the output.

(iii) The liquid water cavity’s absorption spectrum and

that of the water vapor are not the same.

A cheap alternative to generating solar electric power are

Silicon-Film solar cells. The production systems of

Silicon-Film use a continuous in-line process to produce

polycrystalline silicon sheets. The Apex sheet growth

process has continuously progressed and after five design

generations, one sheet can give an annual yield of over 15

MW of 200-mm wide polycrystalline silicon sheet, which

is used to make large-area APx-8 solar cells. These cells

generate over 4 W each and have edge dimensions of

208mm x 208mm. Parallel process systems are often used

to increase the volume of solar cells manufactured.

However, with modern large-area Apex sheet generation

approaches and the development of solar cell process tools,

single-thread process systems can be used to design large

volume production lines for solar cells [7-11].

The greatest verified efficiencies for a number of

photovoltaic cell and module technologies is regularly

recorded in many references. The information has helped

researchers stay aware of the state of the art and to

document their own independent findings in a standardized

manner. All results have to be verified by a recognized test

center before they are included in the tables [12-16].

There are three significant areas relevant to each

photovoltaic device: designed illumination area, aperture

area and total area. The efficiencies of the ‘active area’ are

excluded. Depending on the type of device, a particular

IJCSI International Journal of Computer Science Issues, Vol. 10, Issue 1, No 3, January 2013 ISSN (Print): 1694-0784 | ISSN (Online): 1694-0814 www.IJCSI.org 658

Copyright (c) 2013 International Journal of Computer Science Issues. All Rights Reserved.

http://www.solar-facts-and-advice.com/cadmium-telluride.html

minimum area is an important parameter (800 cm2 for

modules, over 0.05 cm2

for concentrator cells and 1 cm2

for one-sun cells). Cells and modules can be made from a

variety of semiconductors. Also, each semiconductor can

be further categorized into different types (such as thin

film, crystalline and polycrystalline). By provides spectral

information with plotting the external quantum efficiency

(EQE) against the wavelength, with normalization done to

the peak measured value [16-19].

Recent years have seen a dramatic increase in the solar

industry, due to which solar cell and module test solutions

are widely sought after. The solar cell modules are

generally of two types: comprehensive turnkey solutions

and test-system building blocks that must be assembled.

The former are easy to set up but expensive. Furthermore,

the technology used in the turn-key solution is likely to

become outdated quickly, thus requiring an upgrade. With

building blocks though, the system is more reasonably

priced and is easily modified when required. If there is a

need to upgrade to a higher current range or accuracy, only

one relevant block would need to be replaced. Also, sets of

blocks that are useful for a variety of platforms can be

standardized and reused [20,21].

Many automotive facilities are used in the solar cell

manufacturing process for mass production. In thin film

solar cell production especially, a key task is to handle

large size solar cell substrates like the LCD production

system. Robots like the serial manipulator, beam type and

link type robot can be used to handle the large substrate. A

variety of handling robots have been developed over the

years for the manufacturing line for solar cells. With an

increase in the size of the substrate, vibration control and

dynamic analysis assumes greater importance. Since the

weight of the solar cell substrate is at least three times

more than the LCD glass substrate, the position control

needs to be precise, especially considering the vibration of

the substrate and forks. The motion analysis of the beam

type handling robot using Matlab/ SimMechanics was

performed by the authors in a previous research [22 - 24].

In this paper, the design and analysis of a handling robot to

transfer solar cells from the surface of the conveyer to four

main boxes is proposed, with special consideration of the

percentage of efficiency. The paper is divided as follows:

Section I is the introduction. Section II explains the

problem formulation. Section V presents the design of a

prototype by VR. Section IV describes the design of the

controller. Section V explains the simulation of design

with VR and finally conclusion.

2. Problem formulation

A solar cell tester designed previously at University

Malaya’s Power Energy Dedicated Advanced Centre

(UMPEDAC) was used in this work. The structure of the

tester is illustrated in Fig. 1. The tester moves the cells

upon the conveyer till it enters the texting box. The entry is

detected by a photoelectric sensor, after which the tester

box commences work. Each solar cell undergoes a testing

process of 2.5 sec duration. Based on the test results, the

solar cells are divided into four groups.

Fig. 1. Real solar cell tester

The tester calculates efficiency values of the solar cells in

series, which are used to determine the appropriate box the

cells are to be placed in. The robot design thus has two

main objectives: to pick the solar cells from the conveyer

and place them in the suitable box corresponding to the

efficiency value.

3. Design Robot Arm

The robot is designed in two main phases. In the first

phase, the robot is designed in a VR environment and the

controller performance is tested in simulation. VRLM

software was used for the design.The ‘classical objects’

feature in the software was used to create the preliminary

design. The objects were then redrawn using the indexed-

face option to obtain the advanced design. Fig. 2 illustrates

the design of the robot arm in VR.

To design and simulate the movement of the robot, a

trajectory was set for the robot after a number of trials. The

trajectory advances in three stages (Fig. 3): first, the cell is

handled and moved vertically up to a specified point;

second, the cell is moved horizontally to a certain point

and finally the cell is lowered vertically to box A, B, C or

Testing area

Solar cell

Conveyer

part 1

Conveyer

part 2

Controller panel Controller panel

Flash light



D. The robot needs to complete this trajectory in 30 sec.

Due to the thin construction of the solar cell and light

weight (20 gm), it is quite difficult to pick the cells from

the conveyer.

To ensure that each part of the robot was best suited to its

job, a number of robot designs were tested in simulation.

For picking the cells from the conveyer (which requires

accurately capturing them through the conveyer trajectory)

and placing the cells in the appropriate box, the vacuum

technique was used. Two vacuum grippers were used for

this purpose, as is shown in Fig. 3.

Fig.2. Design robot arm in VR environment

Fig. 3. Suction grippers of robot

For high speed up-down vertical movement, the electric

cylinder DNCE with a mechanical linear axis and piston

rod was selected (Fig. 4). The cylinder has low weight and

fast response. The drive component consists of an

electrically driven spindle, which converts the rotary

motion of the motor into the linear motion of the piston

rod. The mechanical interfaces are largely compatible with

the standard cylinder DNC.

The power electronics with positioning controller were

designed as an external field device (IP54). Speed, power

and position can be set independently of each other and up

to 31 travel profiles can be stored. The positioning

controller is suitable for stand-alone operation or can be

externally controlled via an I/O interface or fieldbus. The

movement of the robot in horizontal direction is achieved

using the electric toothed belt axis, drive axis (for

applications with external guide or for easy handling

tasks), plain-bearing guide, toothed belt covered by a steel

band and a flexible motor mounting on all 4 sides of the

axis. The arrangement provides high feed forces, with

speeds up to 5 m/s and acceleration of 50 m/s2. It also

features space-saving position sensing with proximity

sensors in the profile slot suitable for electric axis EGC

with a recirculating ball bearing guide (Fig.5). All the

different parts are assembled virtually and reconstructed in

one frame to achieve the complete robot shown in Fig. 6.

Fig. 4. The electric cylinder DNCE is a mechanical linear axis

with piston

Fig. 5. Electric toothed belt axis

Solar tester

lamps

Solar cell

Tester

conveyor

Boxes

(A-D)

Horizontal bar

Vertical bar

Support frame

Gripper

Photoelectric

Sensors 1 & 2

Output

DNC profile

Magnetic tape Output

measuring system

Reference sensor

Sensor measuring system



Fig. 6. The overall form of robot design.

Program interfaces made by Festo were used to select the

suitable parts of the robot. In the program, application

parameters such as mounting position, mass, stroke and

precision are input and the required process time is

specified. The drive technology can also be preselected.

The required solution package is selected, which are sorted

by motor and axis technology, component utilization, cycle

time or price. The program also provides detailed results

such as motor characteristic curve, dynamic characteristic

values, system data, product data, and parts list.

4. Controller Design and simulation of solar

cell tester

The solar cell tester consists of three main parts: the

tester, conveyer and handling robot. Each part requires its

own control strategy. The controller has to both determine

the working of each part in stand-alone mode and also

govern the interaction of the different parts of the tester.

Taking the environment of the tester into consideration and

the variety of signals that need to handled, a PLC

controller was seen as a suitable option to create the

controller. Two photoelectric sensors are used; one detects

the entry of the solar cell into the testing area (which turns

on the tester flash) and the other to detect the movement of

the robot to start the handling operation. The duration of

robot operation for each cell, from picking the cell and

returning to the initial position, is 10 sec. The time for

testing each cell and obtaining the efficiency value is 10

sec. The rate of testing for the system is thus 360

cells/hour.

The block diagram of the testing process is shown in

Fig.7. The block diagram used for VR simulation is shown

in Fig.8. The diagram shows the blocks used for

implementing all the steps of the tester: movement of the

cells on the conveyer, working of the flash lamp,

measurement of efficiency and the handling operation of

the robot (vertical and horizontal movements, capture and

release of solar cells at the right locations).

Fig. 7. Block diagram of testing process.

Fig. 8. Simulation of solar cell tester in V. R. environment

5. Simulation Result

The simulation of the design was achieved by using

Matlab Ver. 2012a with VR environment to implement the

movement of system by interfacing with Matlab/Simulink.

Fig.9 presents the simulation results by using oscilloscope

display of the control signal for each part of the solar cell

tester.

The operation of the cell testing system commences

with the movement of the solar cell upon the conveyer.

Scope (a) in Fig.9 shows the movement signal for three

cells. As is evident from the signal, the sequence repeats

after every 50 sec. After 10 sec, the first cell arrives at the

tester and the flash tester turns on and off for 5 seconds

each. The sequence is then repeated for the second cell (as

is clear from the tester flash signal in scope (b) of Fig.9).

After 25 sec, the first cell arrives at the capture point to be

picked by the gripper of the robot. The down-up signal for

the gripper is shown in scope (c) in Fig.9. The horizontal

movement of the robot depends on the efficiency of the

cell calculated by the tester. Depending on the box number

that the cell is to be placed in, the amplitude of the

Enter cell to

conveyor

Enter

conveyor

after test

Handling

robot

process

10 sec

Under

testing

process

10 sec



horizontal movement signal (scope (d) of Fig.9) is limited.

The duration of one cycle, from passing the entry point to

being placed in a box is 50 sec. The output of the virtual

model that simulates the movement is shown in Fig. 10.

Fig. 9. (a) Robot arm horizontal movement signal. (b)

Robot arm vertical movement signal. (c) Solar cells

movement signal. (d) Tester flash signal.

Fig. 10. Implement solar cell tester in V.R. model.

6. Conclusion

In this work, the design of a handling robot with vacuum

end-effectors was created to sort solar cells according to

their electrical performance. A PLC was used as the

controller. The motion of the robot and the testing process

were controlled based on detections by photoelectric

sensors. The cells were allotted to suitable boxes based on

their efficiencies. A highly accurate robot trajectory was

obtained, coupled with very accurate position control for

placing the cells inside the boxes. Excellent performance is

obtained in the simulation of the tester in a V. R. model. The

photoelectric sensors were seen to be very effective for

detection of cell entry and robot movement. Future works

can investigate other designs for the robot and look into

different methods of controlling the robot trajectory.

Acknowledgement

Thanks to UMPEDAC for various resources and R&D

project 110775 that funded this research, and to University

of Malaya for the post-doctoral research fellowship.

References [1] Bhushan L, Craig M. Design of a fiber optic based solar

simulator, IEEE Photovoltaic Specialists Conference,

1991, pp. 783-788

[2] Green M. A., Emery K, Hishikawa Y, Warta W. Solar cell

efficiency tables (version 37), progress in photovoltaic:

research and applications, 2011, 19, pp:84–92

[3] Glatthaar, M, Rein, S. Separation of series resistance and

space charge region recombination in crystalline silicon

solar cells from dark and illuminated current–voltage

characteristics. IEEE journal of photovoltaic, Vol. 2, issue

3, 2012, pp. 241-246

[4] Sharma K, Peterson D, Jun B, Hanley J. Reliability

testing of large area 3J space solar cells. 37th IEEE

photovoltaic specialists conference (PVSC), 2011,

pp: 003707 – 003712

[5] Alam A, Upadhyay S, Murthy H, Reddy B, Jana C,

Mohanta K. Reliability evaluation of solar photovoltaic

microgrid. International Conference on Environment and

Electrical Engineering (EEEIC), 2012, pp: 490- 495

[6] Nelson J, The physics of solar cells, ISBN: 978-1-86094-

340-9, May 2003, pp: 384

[7] Wurfel P, physics of solar cells: From Basic Principles to

Advanced Concepts, WILEY-VCH Verlag GmbH & Co.

KGaA, Weinheim, ISBN: 978-3-527-40857-6, 2009,

PP:181

[8] Mohammad S, Suria C, Zurita Z, Faridah J, Kasmiran J.

Portable Wireless CATV Tester Unit with Solar Panel,

IEEE International Conference on Space Science and

Communication,2009, PP: 15-18

[9] Hacke P, Terwilliger K, Smith R, Glick S, Pankow J,

Kempe M, Kurtz S. System voltage potential-induced

degradation mechanisms in PV modules and methods for

test. The 37th IEEE Photovoltaic Specialists Conference

(PVSC 37), 2011, pp: 1-9

[10] Meiqin1 M, Jianhui S, Chang L, Kai1 P,Guorong Z, Ming

D. Research and development of fast field tester for

characteristics of solar array. IEEE electrical and computer

engineering, 2009, pp: 1055-1060

-

c- -d-

-a- -b-



http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=517


http://ieeexplore.ieee.org/search/searchresult.jsp?searchWithin=p_Authors:.QT.Sharma,%20S.K..QT.&newsearch=partialPref

http://ieeexplore.ieee.org/search/searchresult.jsp?searchWithin=p_Authors:.QT.Peterson,%20D..QT.&newsearch=partialPref

http://ieeexplore.ieee.org/search/searchresult.jsp?searchWithin=p_Authors:.QT.Jun,%20B..QT.&newsearch=partialPref

http://ieeexplore.ieee.org/search/searchresult.jsp?searchWithin=p_Authors:.QT.Hanley,%20J..QT.&newsearch=partialPref



[11] Osterwald C. Terrestrial photovoltaic module accelerated

test-to-failure protocol. Technical report, National

Renewable Energy Laboratory 1617 Cole Boulevard,

Golden, Colorado 80401-3393 303-275-3000 •

www.nrel.gov, March 2008, pp:1-16

[12] Warner H, Messenger S, Lorentz, Summers G. On the

need for low energy proton testing of space solar

cells. IEEE 4th World Conference on Photovoltaic Energy

Conversion, Vol. 2, 2006, pp: 1899 - 1902

[13] Shammas E , Brown B, Choset H, New joint design for

three-dimensional hyper redundant robots, IEEE Intl.

Conference on Intelligent Robots and Systems, Nevada,

2003, pp:3594-3599

[14] Rigatos G. Control of robotic systems with flexible

components using hermite polynomial-based neural

networks, Ch-25.book, robot manipulators, new

achievements, Unit of Industrial Automation Industrial

Systems Institute,26504, Greece, 2012, pp:456-486

[15] Rehiara A. Kinematics of Adept Three Robot Arm, ch. 2, at

book, Robot Arms, University of Papua, Indonesia, 2011,

pp: 21-38

[16] Potgieter J, Zyzalo J, Diegel O. Reconfigurable

Mechatronic Robotic Plug-and-Play Controller. Ch:

Cutting Edge Robotics, ISBN 3-86611-038-3, Germany,

July 2005, pp: 771-784

[17] Elfasakhany A, Yanez E, Baylon K, Salgado R. Design and

Development of a Competitive Low-Cost Robot Arm with

Four Degrees of Freedom. Journal of modern mechanical

engineering, 2011, 1, pp: 47-55

[18] Lathrop R, Pfluke K, Novel approaches to benchmarking

solar cell tabbing solder ability. Proceedings of the 26th

European Union Photovoltaic Solar Energy Conference,

2011 , pp: 1-6

[19] Student Guide, Robotics with the Boe-Bot, Ver. 2.2, ISBN

1-928982-03-4, 2011 , pp:1-360

[20] Park C, Park D, Min H. Controller design and motion

simulation of solar cell substrate handling robot in vacuum

environment, 11th International conference on control,

automation and systems, 2011, pp: 1017-1019

[21] Park C, Park D, Min H. Motion simulation model for beam

type solar cell substrate transport robot, The 8th IEEE

International Conference on Ubiquitous Robots and

Ambient Intelligence, 2011, pp:796-799

[22] Wang Z, Gong Z, Wang Y, Wei G. Study on Positioning

Control of Transfer Robot with Solar Cell, The ninth IEEE

international conference on electronic measurement &

instruments, 2009, pp:3,913 3,919

[23] Zhangl X, Panland H, Wul G. Photovoltaic generation and

Its applications in DC-motor, IEEE International

conference on communications, circuits and systems, 2010,

pp: 609-611

[24] Zarafshan P, Ali S, Moosavian A. Manipulation control of

a space robot with flexible solar panels, IEEE/ASME

International conference on advanced intelligent

mechatronics, 2010, pp: 1099-1104

Yousif Ismial Mohammed AL Mashhadany (MIEEE, MIIE) was born in Baghdad 1973. He received the B.Sc. degree in Electrical and Electronic Engineering Department (1995) from the AL-Rasheed College of Engineering and Science / the University of Technology Baghdad / Iraq.M.Sc degree in Control Engineering (1999).and Ph.D degree in Control (2009) from the AL-Rasheed College of Engineering and Science / the University of Technology/ Baghdad / Iraq. Since 2004, he has been working at the University of Anbar – Iraq, as a Lecturer in the Electrical Engineering Department. His research interests include biomedical, robotic and control system. He has more than thirty publishing at journals and International conferences and two books. Now, he has Post-Doctoral research fellows at university of Malaya – UMPEDAC.

Professor Dr. Nasrudin Abd Rahim has been an academician and an active researcher for most of his professional life. Upon graduating with his first degree in 1985, a B.Sc. (Hons.) in Electrical Engineering from the University of Strathclyde in Glasgow, UK, he served briefly in industry, in the capacity of Planning Engineer with a Malaysian telecommunications company. He then entered the world of academia, at the University of Malaya, progressing from Tutor, in 1986, to Lecturer, a year later, and to his doctoral degree in 1988. He was promoted to Associate Professor in 1998, and to Professor in 2003. He was made a Chartered Engineer on 1st January 2000. His professional contribution in imparting and adding to knowledge, and furthering progress in his field, extends beyond the perimeters of the university. At regional and international levels, he has served in various capacities for IEE and IEEE, the most recent has been the Chair of the IEEE Power Engineering Society Motor Sub-Committee Working Group 8 upon his election to the position in 2006. He also has been made the Malaysia SEE Forum Coordinator in 2009. In research, he has led, and co-researched, projects of national and international importance, with funding totalling millions of ringgit. He is a published author of more than 100 refereed journals, 175 articles, and books, and has had more than fifteen PhD candidates graduated. Among many other academic and field-related awards won locally and internationally, is the AUNSEED-Net Research Grant. HE is director for University of Malaya Power Energy Dedicated Advanced Centre (UMPEDAC) from three years.



http://ieeexplore.ieee.org/search/searchresult.jsp?searchWithin=p_Authors:.QT.Warner,%20J.H..QT.&newsearch=partialPref

http://ieeexplore.ieee.org/search/searchresult.jsp?searchWithin=p_Authors:.QT.Messenger,%20S.R..QT.&newsearch=partialPref

http://ieeexplore.ieee.org/search/searchresult.jsp?searchWithin=p_Authors:.QT.Summers,%20G.P..QT.&newsearch=partialPref







design of a pneumatic robotic arm for solar cell tester

Documents